JW Recommendations: Anticipating Your Audience’s Content and Viewing Needs

Blog 3 min read | Mar 22, 2018 | JW Player

Share:

Updates that empower publishers to directly enhance engagement with the video experiences viewers want, curated at scale

Online video continues its explosive growth, and publishers are well positioned to capitalize on increasing viewership. However, they are often limited by the fragmentation of, and competition for, viewer attention in this new landscape.

JW Recommendations is an automated solution that surfaces contextually relevant content from a publisher’s library to its audiences, at scale. Recommendations curates content in real-time and presents stunning experiences that entice viewers to watch more all while providing publishers with complete control. In a two-part blog series, we will cover all the important information about today’s release of the latest JW Recommendations. This updated tool gives publishers what they need to anticipate and deliver the experiences that their viewers desire—deepening audience engagement.

Intelligent recommendations, delivered at scale

Intelligent Recommendations in action: Recommending additional ‘cute animal’ content based on the original video ‘Lost Lion Cub Keeps Crying for Mom’

 

Surfacing relationships between video assets can be a highly manual and time-consuming activity for video content teams. Today’s release addresses this issue by introducing advanced machine learning technologies that enable the JW Recommendations engine to uncover deep content relationships within a publisher’s library of video assets than what was possible before.

And this understanding means a larger quantity of relevant recommendations to viewers. Using semantic similarity techniques, for example, Recommendations understands that a video tagged with “Captain America” is related to many other concepts, including ‘superhero,’ ‘Avengers,’ ‘Marvel,’ and ‘Disney.’ This approach increases both viewer engagement, and the number of recommendations presented—at scale.

And the results are clear: Test campaign results show clickthrough increases up to 13% since semantic similarity has been implemented into JW Recommendations.

Tools for Control: Crafting tailored experiences for audiences

Effortlessly include and exclude content by using our tag-based system

 

Publishers have a new arsenal of tools aimed at better accommodating the output of our Recommendations engine to publisher needs, resulting in more targeted engagement. Recommendations’ ‘Filter by Tag’ now includes Exclusion Rules, which allow the exclusion of videos that are not relevant or appropriate to viewers. By simply typing in one or more content tags, publishers can craft a very specific experience for audiences.

 

Use URL Tag Filtering to modify feeds for advanced use cases

 

Similarly, JW Recommendations now supports ‘URL Tag Filtering,’ providing advanced users the same control and granularity that dashboard users have in including and excluding content by simply modifying the Recommendations feed URL. One of the key new features of URL Tag Filtering is the ability to include and exclude content by the ‘Publish Date.’ This is valuable for publishers who lack evergreen content and want to prioritize viewership of new videos.

These advanced tools mean that Recommendations can be easily fine-tuned by publishers for all use cases and experiences inside or outside the video player.

And this is just the beginning

In our next blog post, which will be published on April 4, we will cover new presentation capabilities that will help create great audience experiences.

To get an interactive view of how our improvements to JW Recommendations can further power your revenue goals, register for our March 29th webinar or schedule time to speak with a video expert.

 

Contact Us